Spinal MNs, which directly control the contraction of skeletal muscle fibers, are located in Rexed lamina IX in the ventral horn of spinal cord (Rexed, 1952). From here, they send their axons out of the CNS in peripheral nerves to innervate the skeletal muscle fibers. Due to their relative accessibility, historically, spinal MNs were used as ‘model neurons’ and provided important insights into general neuronal physiology (Kernell, 2006). Pioneering work by Sir Charles Sherrington, Sir John Eccles, Ragnar Granit, Daniel Kernell, Lord Edgar Adrian, Woodbury, Harry Patton, Brock and others yielded key insights into the function of these neurons (Brownstone, 2006) . Further, spinal MNs are among the first central neurons to be extensively studied using intracellular electrodes (Brock et al., 1952). MNs are unique in that, their function is precisely known: to drive muscle contraction. MN cell bodies are larger than most spinal neurons. Vertebrate MNs are multipolar neurons with extensive dendritic profiles (Kernell, 2006). The heterogeneity of MNs regarding their innervation of distinctive muscle fiber types is reflected by systematic differences in their intrinsic electrical properties and in their susceptibility to degeneration in
13
neurodegenerative conditions, including ALS and ageing (Bakels and Kernell, 1993;
Pun et al., 2006; Saxena and Caroni, 2011).
Spinal MNs can be broadly divided into 3 major classes: alpha, beta and gamma-MNs (Kernell, 2006). These classes can be further subdivided into several different subtypes. Alpha-MNs innervate the force generating extrafusal muscle fibers and can be further classified into FMNTs namely, FF, FR, Fint (FI) and S MNs, based on the motor units they form with the distinct muscle fiber types (Fig. 1). This diversity is essential for orderly and reproducible recruitment of motor units that underlies the gradual build-up (development) of muscle force during movements (Kernell, 2003) (Fig. 1). The discharge properties of MNs are matched to the properties of the muscle fibers they synapse with (Bakels and Kernell, 1993). Knowing the electrophysiological properties of MNs, the type of motor unit it forms and the muscle fiber types it innervates can be predicted with high accuracy (Gardiner, 1993; Zengel et al., 1985). Further, it has been shown that motor nerve activity can profoundly impact muscle fiber phenotype during adult life (Buller et al., 1960; Gordon et al., 1997). Gamma (fusimotor)-MNs innervate intrafusal muscle fibers and are important in maintaining muscle tone (Manuel and Zytnicki, 2011). They can be further subdivided into dynamic and static types based on the type of discharge they elicit at the spindle sensory endings (Bessou et al., 1962a; Bessou et al., 1962b). The beta-MNs are not well characterized and understood, and are thought to innervate both extra and intrafusal muscle fibers and can share characteristics with either FF or S alpha MNs. (Kernell, 2006). MNs can be also classified as extensor and flexor MNs, depending on whether they innervate extensor or flexor muscles. It has been shown that extensor and flexor MNs display different firing profiles, maturation patterns and seem to be incorporated into distinct premotor circuits (Cotel et al., 2009; Tripodi et al., 2011; Vinay et al., 2000).
Spinal MNs have been studied extensively using electrophysiological recordings (Kernell, 2006). The effects of neuromodulatory factors on MN properties are well appreciated (Han et al., 2007; Heckman et al., 2009; Hultborn and Kiehn, 1992;
Muramoto et al., 1996). Further, the initial events of MN development have been extensively studied. For example, molecular pathways involved in the motor neurogenesis, their arrangement into motor columns supplying distinct muscle
14
groups and their organization into motor pools have been well characterized (Bonanomi and Pfaff, 2010; Briscoe et al., 2000; Jessell, 2000; Jurata et al., 2000;
Lee and Pfaff, 2001; Shirasaki and Pfaff, 2002). The processes were further shown to be inherently linked to the establishment of accurate MN-muscle connectivity patterns. However, molecular mechanisms that underlie the functional properties of the distinct MN types (gamma versus alpha or fast versus slow) or those that drive their specification in the first place, remain elusive. Some studies indicate that fast-slow distinction exists in MNs before muscle innervation, but the mechanisms governing the acquisition of these distinctions remain unknown (Rafuse et al., 1996).
Similarly, while maturation of MN presynaptic terminals and possibly some MN type-dependent properties, depend on signals provided by muscle (Chakkalakal et al., 2010; Fox et al., 2007), whether this is also involved in the acquisition of bona fide MN functional type status remains unknown.
Figure 1: Motor neurons and their associated muscle fibers. Spinal MNs are located in the ventral horn of the spinal cord. From here, they send their axons in peripheral nerves to innervate muscle fibers. Spinal MNs can be broadly divided into alpha, beta and gamma types. Alpha-MNs innervate extrafusal muscle fibers (EF) and can be further subdivided into functional MN types (FMNTs)- namely, αFF, αFI, αFR and αS. These FMNTs innervate type IIb, IIx/d, IIa and type I muscle fibers, respectively. Gamma-MNs innervate intrafusal muscle fibers (IF).
15 Fast and slow MNs can be reliably identified by using electrophysiological recordings. For instance, membrane electrical properties can be used to predict the motor unit type (Gardiner, 1993; Zengel et al., 1985). In general, fast MNs have a higher rheobase, lower input resistance and shorter AHP half decay time when compared to slow MNs (Zwaagstra and Kernell, 1980) with AHP half decay times being the best predictors of fast versus slow MN type status (Zengel et al., 1985).
Further, differences in terms of dendritic branching (dendritic bundles in prominent in slow MNs (Gramsbergen et al., 1996; Westerga and Gramsbergen, 1992)), firing behaviour (phasic versus tonic), bistable behaviour (low versus high, plateau potentials above spike threshold in fast versus plateau potentials at or below spike threshold in slow MNs (Lee and Heckman, 1998a; Lee and Heckman, 1998b)), late adaptation (more prominent in fast when compared to slow MNs (Kernell and Monster, 1982)) and NMJ morphology (more complex in fast versus less complex in slow MNs (Kanning et al., 2010)) exist between fast and slow MNs. Further, the fast and slow motor nerve terminals also differ in synaptic vesicle dynamics (Reid et al., 1999). Recent studies have begun to identify putative molecular markers for fast, slow and gamma-MNs: Calca (calcitonin gene-related peptide), Chodl (chondrolectin) for fast MNs, SV2a (synaptic vesicle glycoprotein 2 a) for slow MNs, Err3, Gfra1 (glial cell line derived neurotrophic factor family receptor alpha 1), 5ht1d (5-hydroxytryptamine (serotonin) receptor 1d) and Wnt7a (wingless-related MMTV integration site 7A) for gamma-MNs (Ashrafi et al., 2012; Chakkalakal et al., 2010;
Enjin et al., 2012; Enjin et al., 2010; Friese et al., 2009). Markers like osteopontin and NeuN (neuronal-nuclear antigen) can be used for distinguishing alpha versus gamma-MNs (Misawa et al., 2012). However, particularly markers for fast/slow MNs remain to be verified, and whether any of these play a role in determining fast versus slow, alpha versus gamma-MN properties or type status remains to be addressed. A detailed study of gene signatures specific to fast and slow MNs is currently unavailable. Such a study is interesting in the context of different physiological behaviour of fast and slow MNs and their well-established differential susceptibility towards degeneration in neurodegenerative conditions (Hegedus et al., 2008). Thus, there exists a significant gap in our knowledge regarding FMNTs.
16